Determining investigatory levels of friction with crash modelling

ABSTRACT

Friction is a property of the pavement that provides vehicles with grip to change speed and direction. When the available friction is reduced through the polishing action of traffic, the risk of crash increases. This is especially true when the pavement surface is wet. To reduce the risk of skid-related crashes, pavement friction management programmes may establish thresholds, for example, investigatory levels, for friction based on crash risk analysis. When the friction is at or below this threshold, a site investigation is triggered to determine the cause of the deficiency. Pavement friction management programmes often rely primarily on a visual approach to establish investigatory levels, the results of which are subjective and may vary among analysts. This paper proposes a statistical modelling approach for determining investigatory levels using crash and pavement friction data. For the data used in this paper, the findings show that the proposed approach produced values that were relatively similar in magnitude to the values found in friction policy in Australia, New Zealand, and the United Kingdom.

KEYWORDS:

• Friction
• highway safety management
• friction investigatory levels
• negative binomial regression
• friction demand category

Disclosure statement

No potential conflict of interest was reported by the author(s).